Low-grade systolic murmurs in healthy middle-aged individuals: innocent or clinically significant? A 35-year follow-up study of 2014 Norwegian men

Algorithm for predicting valvular heart disease from heart sounds in an unselected cohort

Objective

This study aims to assess the ability of state-of-the-art machine learning algorithms to detect valvular heart disease (VHD) from digital heart sound recordings in a general population that includes asymptomatic cases and intermediate stages of disease progression.

Methods

We trained a recurrent neural network to predict murmurs from heart sound audio using annotated recordings collected with digital stethoscopes from four auscultation positions in 2,124 participants from the Tromsø7 study. The predicted murmurs were used to predict VHD as determined by echocardiography.

Results

The presence of aortic stenosis (AS) was detected with a sensitivity of 90.9%, a specificity of 94.5%, and an area under the curve (AUC) of 0.979 (CI: 0.963-0.995). At least moderate AS was detected with an AUC of 0.993 (CI: 0.989-0.997). Moderate or greater aortic and mitral regurgitation (AR and MR) were predicted with AUC values of 0.634 (CI: 0.565-703) and 0.549 (CI: 0.506-0.593), respectively, which increased to 0.766 and 0.677 when clinical variables were added as predictors. The AUC for predicting symptomatic cases was higher for AR and MR, 0.756 and 0.711, respectively. Screening jointly for symptomatic regurgitation or presence of stenosis resulted in an AUC of 0.86, with 97.7% of AS cases (n = 44) and all 12 MS cases detected.

Conclusions

The algorithm demonstrated excellent performance in detecting AS in a general cohort, surpassing observations from similar studies on selected cohorts. The detection of AR and MR based on HS audio was poor, but accuracy was considerably higher for symptomatic cases, and the inclusion of clinical variables improved the performance of the model significantly.

Beyond Heart Murmur Detection: Automatic Murmur Grading From Phonocardiogram

OBJECTIVE

Murmurs are abnormal heart sounds, identified by experts through cardiac auscultation. The murmur grade, a quantitative measure of the murmur intensity, is strongly correlated with the patient's clinical condition. This work aims to estimate each patient's murmur grade (i.e., absent, soft, loud) from multiple auscultation location phonocardiograms (PCGs) of a large population of pediatric patients from a low-resource rural area.

METHODS

The Mel spectrogram representation of each PCG recording is given to an ensemble of 15 convolutional residual neural networks with channel-wise attention mechanisms to classify each PCG recording. The final murmur grade for each patient is derived based on the proposed decision rule and considering all estimated labels for available recordings. The proposed method is cross-validated on a dataset consisting of 3456 PCG recordings from 1007 patients using a stratified ten-fold cross-validation. Additionally, the method was tested on a hidden test set comprised of 1538 PCG recordings from 442 patients.

RESULTS

The overall cross-validation performances for patient-level murmur gradings are 86.3% and 81.6% in terms of the unweighted average of sensitivities and F1-scores, respectively. The sensitivities (and F1-scores) for absent, soft, and loud murmurs are 90.7% (93.6%), 75.8% (66.8%), and 92.3% (84.2%), respectively. On the test set, the algorithm achieves an unweighted average of sensitivities of 80.4% and an F1-score of 75.8%.

CONCLUSIONS

This study provides a potential approach for algorithmic pre-screening in low-resource settings with relatively high expert screening costs.

SIGNIFICANCE

The proposed method represents a significant step beyond detection of murmurs, providing characterization of intensity, which may provide an enhanced classification of clinical outcomes.

The Oslo Ischaemia Study: cohort profile

PURPOSE

The Oslo Ischaemia Study was designed to investigate the prevalence and predictors of silent coronary disease in Norwegian middle-aged men, specifically validating exercise electrocardiography (ECG) findings compared with angiography. The study has been important in investigating long-term predictors of cardiovascular morbidity and mortality, as well as investigating a broad spectrum of epidemiological and public health perspectives.

PARTICIPANTS

In 1972-1975, 2014 healthy men, 40-59 years old, were enrolled in the study. Comprehensive clinical examination included an ECG-monitored exercise test at baseline and follow-ups. The cohort has been re-examined four times during 20 years. Linkage to health records and national health registries has ensured complete endpoint registration of morbidity until the end of 2006, and cancer and mortality until the end of 2017.

FINDINGS TO DATE

The early study results provided new evidence, as many participants with a positive exercise ECG, but no chest pain ('silent ischaemia'), did not have significant coronary artery stenosis after all. Still, they were over-represented with coronary disease after years of follow-up. Furthermore, participants with the highest physical fitness had lower risk of cardiovascular disease, and the magnitude of blood pressure responses to moderate exercise was shown to influence the risk of cardiovascular disease and mortality. With time, follow-up data allowed the scope of research to expand into other fields of medicine, with the aim of investigating predictors and the importance of lifestyle and risk factors.

FUTURE PLANS

Recently, the Oslo Ischaemia Study has been found worthy, as the first scientific study, to be preserved by The National Archives of Norway. All the study material will be digitised, free to use and accessible for all. In 2030, the Oslo Ischaemia Study will be linked to the Norwegian Cause of Death Registry to obtain complete follow-up to death. Thus, a broad spectrum of additional opportunities opens.

Aortic valve stenosis awareness in Austria-results of a nationwide survey in 1001 subjects

Despite the prognostic significance of severe aortic valve stenosis, knowledge is limited in the general population. To document the status quo for Austria, knowledge about valvular heart disease/aortic valve stenosis was documented in 1001 participants >60 years of age. 6.7% of respondents were knowledgeable of aortic valve stenosis, with 1.6% being concerned about the condition (24.1% cancer, 18.8% Alzheimer’s disease, 15.1% stroke). 29.5% were familiar with valvular heart disease (76.7% heart attack, 36.9% stroke). Only 1/3 reported auscultation by their general practitioner (GP) at least every third visit. Typical symptoms of aortic valve stenosis were likely to be reported by 50%. After exposure to further information on aortic valve stenosis, only 20% reported to be more concerned and ready to obtain more disease-related information. Awareness of surgical and catheter-based treatment options was claimed by 77% of respondents. Awareness campaigns on valvular heart disease are warranted to improve patient care in Austria.

Patient screening for early detection of aortic stenosis (AS)-review of current practice and future perspectives

In Europe, approximately one million people over 75 years suffer from severe aortic stenosis (AS), one of the most serious and most common valve diseases, and this disease burden is increasing with the aging population. A diagnosis of severe symptomatic AS is associated with an average life expectancy of 2-3 years and necessitates a timely valve intervention. Guidelines for valve replacement therapy have been established but only a proportion of patients with symptomatic AS actually receive this life-saving treatment. The decision for valve intervention in asymptomatic patients with severe AS is often more challenging and likely results in fewer patients receiving treatment in comparison to their symptomatic counterparts. This article reviews the epidemiology and clinical manifestations of AS, the associated economic burden of AS to the healthcare system, the diagnosis of AS and the possible mechanisms for the introduction of routine screening in elderly patients. Elderly patients typically visit healthcare providers more frequently than younger patients, thereby providing increased opportunities for ad hoc AS screening and this, along with raising patient awareness of the symptoms of AS, has the potential to result in the earlier diagnosis and treatment of AS and increased patient survival.

Triglycerides-diabetes association in healthy middle-aged men: modified by physical fitness? A long term follow-up of 1962 Norwegian men in the Oslo Ischemia Study

AIMS

To examine the impact of physical fitness (PF) on the association between fasting serum triglycerides (FTG) and diabetes risk and whether temporal changes in FTG predict diabetes risk in healthy middle-aged men.

METHODS

FTG and PF (bicycle exercise test) were measured in 1962 men aged 40-59 years in 1972-1975 (Survey 1) and repeated in 1387 still healthy men on average 7.3 years later (Survey 2). Diabetes was diagnosed according to WHO 1985-criteria.

RESULTS

During 35 years follow-up 202/1962 (10.3%) men developed diabetes. Compared with the lowest, the upper FTG tertile had a 2.58-fold (95% CI: 1.81-3.74) diabetes risk adjusted for age, fasting blood glucose and maternal diabetes, and a 2.29-fold (95%CI: 1.60-3.33) when also adjusting for PF. Compared with unchanged (±25%) FTG levels (n=664), FTG reduction of more than 25% (n=261) was associated with 56% lower (0.44; 95% CI: 0.24-0.75) diabetes risk, while FTG increase of more than 25% (n=462) was associated with similar risk. These associations were unchanged when adjusted for PF and PF change.

CONCLUSIONS

High FTG-levels predicted long-term diabetes risk in healthy middle-aged men, and the association was only modestly weakened when adjusted for PF. A reduction in FTG was associated with decreased diabetes risk.